CN117666547A

CN117666547A - Recognition method for entering narrow channel of robot, chip and robot

Info

Publication number: CN117666547A
Application number: CN202210980334.4A
Authority: CN
Inventors: 徐松舟; 欧兆锐; 周和文; 黄惠保; 孙明; 陈卓标
Original assignee: Zhuhai Amicro Semiconductor Co Ltd
Current assignee: Zhuhai Amicro Semiconductor Co Ltd
Priority date: 2022-08-16
Filing date: 2022-08-16
Publication date: 2024-03-08

Abstract

The invention discloses a recognition method for a robot to enter a narrow channel, a chip and the robot, wherein an execution main body of the recognition method is a robot provided with a ranging sensor; the identification method comprises the following steps: step S1, when a robot detects an obstacle by using a ranging sensor, the robot plans an obstacle detouring path leading to a set target position, walks along the obstacle detouring path, and acquires a reference contour line segment through point cloud data acquired by the ranging sensor; and S2, under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the robot detects that the walking direction of the robot changes from the reference obstacle detouring direction to the direction pointing to the preset target position from the current position and detects that the distance between the two obstacle detouring points is in the preset distance range, the robot recognizes the behavior as entering a narrow road.

Description

Recognition method for entering narrow channel of robot, chip and robot

Technical Field

The invention relates to the technical field of mobile robots, in particular to a method for identifying a robot entering a narrow channel based on robot obstacle detouring, a chip and a robot.

Background

In the working environment of the sweeping robot, a working area framed by walls exists, and a non-working area is framed between two walls, wherein the non-working area generally refers to a channel with a narrow width, namely a narrow channel for short, and an inlet of the narrow channel and an outlet of the narrow channel are both arranged as narrow channels, so that the narrow channel is divided by wall barriers or divided into a plurality of room areas with narrow channels in the working environment of the sweeping robot; wherein, two sides of the narrow channel are respectively the contour lines of two barriers; and when setting the moving path, the sweeping robot is generally set to one point to be viewed. In order to facilitate the movement of the sweeping robot to a target location in another room area, the path of movement may be set to pass through a narrow aisle.

In the process of carrying out instant positioning and synchronous navigation by using point cloud data detected by line laser, the robot generally autonomously plans a navigation path according to a set task, and after the robot detects an obstacle or a narrow channel in the travelling process, the robot keeps travelling according to the outline of the obstacle in order to avoid the obstacle, and the calculated outline of the obstacle has errors, so that the robot can not pass through some narrow channels easily, and the navigation success rate is reduced.

Disclosure of Invention

In order to solve the problem that a robot accurately identifies a narrow channel, the invention provides an identification method for a robot to enter the narrow channel, a chip and the robot. The specific technical scheme is as follows:

an identification method for a robot entering a narrow channel, wherein an execution main body of the identification method is a robot provided with a ranging sensor; the identification method comprises the following steps: step S1, when a robot detects an obstacle by using a ranging sensor, the robot plans an obstacle detouring path leading to a set target position, walks along the obstacle detouring path, and acquires a reference contour line segment through point cloud data acquired by the ranging sensor; and S2, under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the robot detects that the walking direction of the robot changes from the reference obstacle detouring direction to the direction pointing to the preset target position from the current position and detects that the distance between the two obstacle detouring points is in the preset distance range, the robot recognizes the behavior as entering a narrow road.

Further, the step S2 further includes: under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the distance between the two obstacle detouring points is detected to be in a preset distance range, identifying a channel in which the two obstacle detouring points are positioned as a narrow channel; the direction of the current position pointing to the set target position in step S2 is pointing to the passable area inside the channel where the two obstacle detouring points are located.

Further, the robot configures a reference obstacle detouring direction into an outer contour line trend formed by obstacles where two obstacle detouring points are respectively located so as to form an extending direction of the obstacle detouring path before the robot enters a narrow road; the robot sets the searching direction of the reference contour line segment as the vertical direction of the reference contour line segment, and forms the width direction of the channel where the two obstacle detouring points are located; the reference contour line segment corresponds to an obstacle where one obstacle detouring point is located, and forms one boundary line of the channel where the two obstacle detouring points are located.

Further, in the process that the robot walks along the obstacle detouring path, if the robot adjusts the walking direction of the robot to the direction pointing to the set target position from the current position, and detects that the distance between the two obstacle detouring points is within the preset distance range, the robot determines that the robot starts to enter a narrow opening, the narrow opening is a gap formed between the obstacles where the two obstacle detouring points are respectively located, the gap is located in the passable area, and the direction of the current position pointing to the set target position is configured to be different from the reference obstacle detouring direction by the robot.

Further, between the step S1 and the step S2, further includes: after the robot detects one obstacle and before two obstacle detouring points are extracted in the searching direction of the reference contour line segment of the other obstacle, the robot keeps walking along the obstacle detouring path, and the current extending direction of the obstacle detouring path is set as a historical obstacle detouring direction; and after the robot acquires the reference contour line segment in the process of walking along the obstacle detouring path, if the robot detects that the extending direction of the obstacle detouring path from the current position is not the direction from the current position to the set target position, stopping to walk along the obstacle detouring path, and adjusting the walking direction to the direction from the current position to the set target position so as to enable the robot to start entering the channel where the two obstacle detouring points are located.

Further, when the robot detects a first obstacle by using a ranging sensor, planning an obstacle detouring path leading to a set target position by the robot, and then walking along the obstacle detouring path to realize walking along the contour line of the first obstacle until the robot walks to a gap formed between the first obstacle and a second obstacle and/or collides with the second obstacle, marking the walking direction of the robot as the historical obstacle detouring direction, wherein a reference contour line of the second obstacle has a first end point and a second end point; then the robot acquires a reference contour line segment of the second obstacle, marks the direction of the first end point pointing to the second end point as a direction pointing to a set target position from the current position, and marks the direction of the second end point pointing to the first end point as an extending direction of the history obstacle detouring direction; the robot configures an extension direction of the history obstacle detouring direction as the reference obstacle detouring direction.

Further, the extending direction of the passable area in the channel where the two obstacle detouring points are located is parallel to the reference contour line segment of the second obstacle, the reference contour line segment of the second obstacle is perpendicular to the searching direction, the passable area in the channel where the two obstacle detouring points are located is communicated with the set target position, and one obstacle detouring point is located on the reference contour line segment of the second obstacle.

Further, the robot collects point cloud data through the ranging sensor, the point cloud data being position information configured to reflect an obstacle detected from the ranging sensor; the robot then fits the acquired point cloud data into the contour line of the obstacle to represent the local contour of the detected obstacle or the obstacle envelope; the reference contour line segment belongs to a contour line which is subjected to fitting processing; when the robot marks the extending direction of the passable area in the channel where the two obstacle detouring points are located as the preset channel direction, the two obstacle detouring points are respectively positioned on the contour line of the first obstacle in the preset channel direction and the contour line of the second obstacle in the preset channel direction, or the two obstacle detouring points are respectively positioned on the first obstacle and the second obstacle.

Further, when the robot detects the first obstacle, the position where the robot walks along the obstacle detouring path and the predetermined target position are separated on two sides of the first obstacle, then the robot marks one side where the position where the robot walks along the obstacle detouring path is located as a first side, marks one side where the predetermined target position is located as a second side, the robot configures a first end point of a reference contour line segment of the second obstacle to be located on the first side of the second obstacle or the first side of the first obstacle, and the robot configures a second end point of the reference contour line segment of the first obstacle to be located on the second side of the second obstacle or the second side of the first obstacle; and then the robot marks the profile line trend formed by connecting the profile line of the first side of the first obstacle and the profile line of the first side of the second obstacle as the profile line trend of the outer profile line formed by the obstacles where the two obstacle detouring points are respectively positioned.

Further, when the robot detects the first obstacle first by using the ranging sensor, the robot has collided with the first obstacle; when the robot walks to a gap formed between the first obstacle and the second obstacle, the robot collides with the second obstacle; the robot sets the preset distance range to be greater than or equal to the body width of the robot, and the upper limit value of the preset distance range is the sum of the body width of the robot and the preset redundancy amount.

A chip storing program code which when executed realizes the steps of the method for identifying the entrance of the robot into the narrow channel.

The robot is provided with the ranging sensor, and the chip is used for controlling the robot to detect the obstacle by using the ranging sensor and obtain the corresponding contour line and the obstacle detouring point, so that the robot can be conveniently identified to start entering the narrow road.

The method has the beneficial technical effects that when the robot walks along the obstacle detouring path planned in advance, the robot aims at the problem that a narrow channel which is formed between two obstacles and is just larger than the width of a machine body by a smaller clearance distance (1 cm) is easy to misjudge as non-passing, the contour line segment related to the passing factor of a gap formed between the two obstacles is obtained and used for guiding the robot to adjust the obstacle detouring direction before and after entering the gap, and when the obstacle detouring direction is changed, the distance between two obstacle detouring points searched in the searching direction of the contour line segment is combined, the channel where the two obstacle detouring points are located is identified as the narrow channel, and the condition that the robot starts to enter a narrow channel opening but does not enter the obstacle detouring path planned in advance is determined. Therefore, the narrow channel for the robot to pass is accurately distinguished in the environment where the sweeping robot, the mower or the mobile toy is located, the problem that the width of the narrow channel is small enough to easily misjudge the grid area inside the narrow channel or the opening of the narrow channel as an obstacle blocking the robot to pass is solved, smooth switching from the action of bypassing the obstacle by the robot to the action of entering the narrow channel formed between two obstacles is also realized, and the robot can accurately pass through the narrow channel or the narrow channel according to laser data, so that the smoothness and success rate of passing through the narrow area under the laser navigation condition are improved.

No matter how frequently the robot collides with the wound obstacle before entering the channel where the two obstacle detouring points are located, the walking direction of the robot is changed from the reference obstacle detouring direction to the direction pointing to the preset target position from the current position (namely, the previous obstacle detouring direction is different from the current obstacle detouring direction), and when the distance between the two obstacle detouring points is in the preset distance range, the robot recognizes the channel where the two obstacle detouring points are located as a narrow channel, so that the robot is smoother in the process of navigating from the current position to the preset target position or the navigation path is shorter, and the navigation efficiency of the robot is improved.

Drawings

Fig. 1 is a flowchart of a method for identifying a narrow lane based on obstacle detouring by a robot according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a movement track of a robot before and after entering and exiting a narrow crossing according to another embodiment of the present invention, wherein a black rectangle above the drawing is a second obstacle, and a black rectangle below the drawing is a first obstacle.

Fig. 3 is a schematic view of a robot changing the obstacle detouring direction according to yet another embodiment of the invention, wherein a black rectangle above the illustration is a second obstacle.

Detailed Description

The following describes the technical solution in the embodiment of the present invention in detail with reference to the drawings in the embodiment of the present invention. For further illustration of the various embodiments, the invention is provided with the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments and together with the description, serve to explain the principles of the embodiments. With reference to these matters, one of ordinary skill in the art will understand other possible embodiments and advantages of the present invention. A process or method depicted as a flowchart. Although a flowchart depicts steps as a sequential process, many of the steps may be implemented in parallel, concurrently, or with other steps. Furthermore, the order of the steps may be rearranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figures. The processes may correspond to methods, functions, procedures, subroutines, and the like.

For narrow roads, point cloud data (radar points) collected by the laser sensor disclosed in the embodiment are all represented by discrete coordinates, the discrete coordinates are particularly used for representing the positions of the obstacles, most of the discrete coordinates are needed to be screened and grouped, then the grouped point cloud data are fitted to obtain a fitting straight line or a fitting curve, then an obstacle envelope is obtained through the coordinate points of the fitting curve according to the proportional relation between world coordinates and image coordinates, and the obstacle envelope is displayed in a grid map in real time (the point cloud data are distributed around the obstacle in a surrounding manner), so that the laser detection distance from the obstacle to the edge of the robot is corresponding, and the position and the outline of the obstacle can be represented; when the line laser sensor is used for collecting point cloud data (radar points), as the line laser sensor projects a straight line, the distance of a low obstacle in front of the robot can be measured, the distance can be converted into point cloud data of the obstacle, the point cloud data can be fitted into a straight line segment, and two end points of the straight line segment are configured as reference points for obstacle detouring of the robot in some embodiments and can be regarded as obstacle detouring points; because errors accumulated by the sensors of the robot, different obstacle materials and different light environment interference, an obstacle envelope curve (the peripheral envelope contour of an obstacle, a fitting curve or the connection of a plurality of fitting curves) obtained by using point cloud data acquired by a laser sensor cannot fit an actual obstacle position, the robot excludes a narrow road opening from a passable area of the robot when searching a path, and even if the side length of a grid is set to be moderate, all grids occupied by the narrow road opening with smaller width are marked as obstacle grids, so that the errors of the obstacle envelope curve can cause misjudgment of the robot on the narrow road opening; the robot cannot search for a path into the narrow road junction, resulting in the robot failing to pass through the narrow road. For example, when the sweeping robot cleans, if the sweeping robot moves to a narrow road condition between two walls or the width formed by two low obstacles is slightly larger than the width of the robot body, the sweeping robot actively avoids the obstacle when touching the obstacle, the sweeping robot can avoid the obstacle by greatly turning, and after the collision sensor of the sweeping robot detects the collision, the sweeping robot can rotate to the opposite side and then move around the obstacle, but the sweeping robot does not try to enter the narrow road, so that the full-coverage cleaning can not be realized, the success rate of the robot navigation is reduced, the sweeping robot is easy to be trapped in the middle of four desk legs, and the sweeping robot can not autonomously get rid of the obstacle.

In order to enable the navigation of a robot to smoothly bypass an obstacle and enter a narrow road, the invention discloses a narrow road identification method based on obstacle bypassing based on the adjustment of obstacle bypassing trend direction, and an execution main body of the narrow road identification method is a robot provided with a ranging sensor, wherein the ranging sensor can be a linear laser sensor, and position information reflected by the obstacle is acquired by emitting linear laser to the outside; the line laser sensor comprises a multi-line laser radar and a single-line laser radar, wherein the single-line laser radar refers to a radar with a single line from a laser source, is applied to the field of robots and can help the robots to avoid obstacles, and the line laser sensor has the advantages of high scanning speed, high resolution and high reliability. In particular, a cleaning robot that walks on the ground surface, a mowing robot that walks on a lawn area with a narrow passage defined by a boundary line, a floor scrubber, a security inspection robot, and the like, the present embodiment is not limited as to the type of the main body to which the narrow passage recognition method is applied. The robot may be further provided with an inertial sensor (including but not limited to an odometer for measuring a walking distance, a collision sensor for detecting a collision state with an obstacle, and a gyroscope for measuring a rotation angle of the robot body), or a vision sensor (any type of depth information acquisition device may be adopted, including but not limited to a monocular camera, a binocular camera, etc. cameras) to detect two-dimensional point cloud data of the surrounding environment, a two-dimensional point cloud map may be constructed in time, and the number of sensor installation on the robot body may be one or more.

Referring to fig. 1, the method for identifying the narrow channel includes: step S1, when the robot detects an obstacle by using a ranging sensor, the robot plans an obstacle detouring path leading to a set target position, walks along the obstacle detouring path, and acquires a reference contour line segment by using point cloud data acquired by the ranging sensor. Before the robot executes step S1, the robot will draw a navigation path using heuristic search algorithm rules, including a navigation path searched out in a map using an a-algorithm, and the predetermined target position is the end point of the navigation path; after planning the navigation path, the robot walks along the navigation path, when the robot detects an obstacle by using the ranging sensor, such as detecting the obstacle for the first time, even when the robot collides with the obstacle, the robot starts to draw a navigation path bypassing the currently detected obstacle from the current position or a preset starting point by using heuristic search rules, marks the navigation path as a barrier-bypassing path, and the set target position is the end point of the barrier-bypassing path. Specifically, before the robot starts to walk along the obstacle detouring path, whether the robot collides with an obstacle or not, the current position or the preset starting position of the robot is taken as a searching starting point, the searching starting point is pointed to the direction of the preset target position according to the direction of the searching starting point, the direction of the searching starting point is pointed to the preset target position target in fig. 1 and 2, during the process of planning the path by using a heuristic searching algorithm, a free grid (a grid area representing a passable area in a grid map) is searched in the neighborhood of the searching starting point as a path node so that the path node is not positioned on the obstacle occupation area, wherein the robot searches a plurality of path nodes in the neighborhood of the searching starting point, the path nodes are all free grids, and each path node corresponds to one initial searching direction; then, continuously updating each path node into a new searching starting point, searching a free grid in the neighborhood of each searching starting point according to the direction that the searching starting point points to the same set target position, repeating the steps until the set target position is searched in a grid map, and simultaneously connecting the searched grids into a plurality of navigation paths or obstacle detouring paths according to the sequence of searching; the robot correspondingly searches a navigation path or the obstacle detouring path in each initial searching direction, the width of a grid occupied by each navigation path or obstacle detouring path in the path width direction is larger than the width of a robot body, and each navigation path or obstacle detouring path is a passable path which can extend from the current position of the robot to the set target position. Preferably, the neighborhood may be a grid area consisting of a 4 neighborhood, an 8 neighborhood or a 12 neighborhood centered on a grid where a current position of the robot (a body center position of the robot) is located, and may be a range of 10, 15, 20 or 30 grids in four directions of front, back, left and right of the body center position of the robot, wherein the number of grids 10, 15, 20 or 30 is only exemplified.

In the step S1, the robot generally takes the position of the first collision to one obstacle as a search starting point, then, when the robot collides with the other obstacle in the process of walking along the obstacle detouring path planned at the search starting point, the robot can also be regarded as detecting the other obstacle, then, the point cloud data collected by the ranging sensor acquires a reference contour line, which is a fitting result of the point cloud data reflected from the other obstacle (a part of the line segment from the envelope line of the other obstacle and can overlap with the current position of the robot), so as to explore a gap or a channel formed between the two detected obstacles, and in some embodiments, the reference contour line can be a fitting line calculated based on the point cloud data collected by the robot in advance (such as a contour from the first collision to the obstacle) or a line parallel to the fitting line, and the robot may not detect or collide with the other obstacle, but the robot can predict that a gap may be formed between the two different obstacles; thus, the robot regards the reference contour line segment as the boundary line where one end of the gap is located or the boundary line where one side of the channel is located.

And S2, under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the walking direction of the robot is detected to be changed from the reference obstacle detouring direction to the direction from the current position to the preset target position and the distance between the two obstacle detouring points is detected to be in the preset distance range, the robot recognizes the behavior as entering the narrow road, at the moment, the robot enters the narrow road from the outside of the narrow road, or the robot is positioned in front of the entrance of the channel where the two obstacle detouring points are positioned, the walking direction of the robot is a passable area leading to the inside of the channel where the two obstacle detouring points are positioned, and the robot can twist left and right. Specifically, after the robot acquires the reference contour line segment in step S1, the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, and may extract another obstacle detouring point along the searching direction on the basis of selecting one obstacle detouring point, where the obstacle detouring points are end points of the fitted contour line segment and are two end points closest to the two reference contour line segments, so as to form obstacle detouring points on the left and right sides of the robot; then, the robot can continue to search a new obstacle-surrounding path from the current position, then bypass an obstacle where one obstacle-surrounding point is located along the new obstacle-surrounding path or adjust the walking direction to linearly pass through a gap formed between the two extracted obstacle-surrounding points, the obstacle-surrounding direction formed by the robot is different from the obstacle-surrounding direction formed in the step S1 so as to deviate from the original obstacle-surrounding trend, and if the distance between the two extracted obstacle-surrounding points is detected to be in the preset distance range, the current start of the robot to enter the narrow road is determined. Preferably, the preset distance range is greater than a body width of the robot, which is a body diameter thereof when the outer shape of the robot is circular.

When the robot detects that the distance between the two obstacle detouring points is in the preset distance range, the robot recognizes the channel where the two obstacle detouring points are located as a narrow channel, and further when the robot detects that the walking direction of the robot is changed from the reference obstacle detouring direction to the direction pointing to the preset target position from the current position, the robot is determined to enter the narrow channel. The robot configures the direction of the current position pointing to the set target position to point to a passable area in the channel where the two obstacle detouring points are located, the current position is a path node where the obstacle detouring path planned in the step S1 passes, and the obstacle detouring trend of the robot at the path node changes. Preferably, the preset distance range is greater than or equal to the body width of the robot; when the width of the entrance of the channel where the two obstacle detouring points are located (namely, the gap formed between the two obstacle detouring points) is larger than or equal to the width of the robot body, the channel where the two obstacle detouring points are located is determined to allow the robot to enter, and when the robot walks to one of the obstacle detouring points or the vicinity thereof to collide with a new obstacle, the robot can adjust the walking direction of the robot to the direction from the current position to the predetermined target position, and the adjusted direction can be the direction of pointing the predetermined target position by a straight line or the direction of bending through the channel where the two obstacle detouring points are located to extend to the predetermined target position, and then when the walking direction of the robot is detected to change from the reference obstacle detouring direction to the direction pointing to the predetermined target position from the current position, and the distance between the two obstacle detouring points is detected to be in the preset distance range, the robot recognizes the robot to enter the narrow channel so that the robot smoothly enters and passes through the narrow channel. On the other hand, if it is detected that the width of the entrance of the channel where the two obstacle detouring points are located (i.e. the gap formed between the two obstacle detouring points) is smaller than the width of the robot body, it is determined that the channel where the two obstacle detouring points are located does not allow the robot to enter, then the robot may resume using the obstacle detouring path in step S1 or to avoid the obstacle, or adjust its walking direction to the opposite direction of the direction from the current position to the predetermined target position, or adjust its walking direction to the contour of the obstacle extending from the current position to the latest collision, so that the robot does not enter the channel where the two obstacle detouring points are located, plan to form a new obstacle detouring path, keep walking along the contour of the obstacle, and detouring the new obstacle.

In summary, when the robot walks along the obstacle detouring path planned in advance, the robot is easy to misjudge as an unviewable narrow road which is formed between two obstacles and is just larger than the width of the machine body by a smaller gap distance (1 cm), the contour line segment related to the passing factor of the gap formed between the two obstacles is obtained to guide the robot to adjust the obstacle detouring direction before and after entering the gap, and when the obstacle detouring direction is changed, the distance between two obstacle detouring points extracted in the searching direction of the contour line segment is combined, the channel where the two obstacle detouring points are located is identified as the narrow road, and the condition that the robot starts to enter a narrow road opening but does not enter the obstacle detouring path planned in advance is determined. Therefore, the narrow channel for the robot to pass is accurately distinguished in the environment where the sweeping robot, the mower or the mobile toy is located, the problem that the width of the narrow channel is small enough to easily misjudge the grid area inside the narrow channel or the opening of the narrow channel as an obstacle blocking the robot to pass is solved, smooth switching from the action of bypassing the obstacle by the robot to the action of entering the narrow channel formed between two obstacles is also realized, and the robot can accurately pass through the narrow channel or the narrow channel according to laser data, so that the smoothness and success rate of passing through the narrow area under the laser navigation condition are improved.

On the basis of the embodiment, the robot may mark index numbers for grids where the searched corresponding one of the paths is located in order according to the direction in which the current position points to the predetermined target position in the process of searching each path leading to the predetermined target position in the grid map, and set the extending direction of the grid corresponding to the index numbers sorted from small to large as the walking direction of the robot on the searched corresponding one of the paths to determine the walking direction of the robot on the searched path (including the obstacle detouring path). Preferably, when the current position is required to pass through the narrow road in the process of pointing to the predetermined target position, the direction of pointing to the predetermined target position in step S2 may be the direction from the entrance of the narrow road to the exit of the narrow road, so as to ensure the accessible area of the interior of the channel where the two obstacle detouring points are located, and the walking direction of the robot entering the narrow road may be determined according to the index number of the grid closest to the entrance of the narrow road. As shown in fig. 2, before a robot enters a gap formed between two obstacles above and below, collision is performed on the obstacle below the upper diagram at a start position, the start position is marked as a searching starting point, an index number (1) is marked, then the index number (2) and the index number (3) are marked on searched path nodes along the boundary line of the obstacle below the diagram, and at the position corresponding to the index number (3), the walking direction of the robot is changed to face the inside of the gap and point to a target position, wherein the position corresponding to the index number (3) is the position closest to the gap, and is also the position of the robot for changing the obstacle-surrounding direction; after the robot determines to enter the gap, the index number (4), the index number (5) and the index number (6) are marked at the searched path nodes in sequence according to the principle of the channel where the gap is located until the target position is searched, and the path searching algorithm comprises heuristic searching algorithms including but not limited to an A-type algorithm, a D-type algorithm and the like, so that the searched path can avoid the illustrated obstacle.

As an embodiment, the robot configures the reference obstacle detouring direction as an outer contour line trend formed by obstacles where two obstacle detouring points are respectively located, so as to form an extending direction of the obstacle detouring path before the robot enters the narrow road; wherein, each obstacle-surrounding point corresponds to an obstacle, the robot can be regarded as an obstacle where the obstacle-surrounding point is located by one obstacle-surrounding point, the reference obstacle-surrounding direction can be shown as a direction extending along a connecting line of the outline of the outer sides of the two rectangular obstacles shown in fig. 2, and can be understood as a curve surrounding the outer sides of the two rectangular obstacles shown in fig. 2 and the trend is close to the trend of the outline of the two obstacles; when the robot enters a gap (opening) between two obstacle detouring points, the robot detours around the obstacle where the two obstacle detouring points are respectively located or passes through the gap formed between the obstacles where the two obstacle detouring points are respectively located. On the other hand, in this embodiment, the robot sets the search direction of the reference contour line segment as the vertical direction of the reference contour line segment, so as to form the width direction of the channel where the two obstacle detouring points are located; the reference contour line segment corresponds to an obstacle where one obstacle detouring point is located, corresponds to an obstacle above in fig. 2, the line segment AB is the reference contour line segment, may be a fitting line segment passing through a corner point below and to the left of the obstacle above in fig. 2, and does not necessarily belong to an actual contour line of the obstacle, but can form one boundary line of the channels where the two obstacle detouring points are located, that is, the boundary line of the narrow channel identified in the foregoing embodiment, where the line segment CD shown in fig. 2 is regarded as being perpendicular to the line segment AB, and the points C and D are two obstacle detouring points respectively, and the obstacle detouring point C may be one corner point located on the obstacle (black rectangle) shown in fig. 2, or may be one contour point located on the obstacle in the direction of the perpendicular AB to fit the contour and distribution position of the obstacle. The two obstacle detouring points are points on a reference contour line segment of an obstacle nearest to the left side and the right side of the robot in the process of executing the identification method, and can be understood as two end points with the smallest distance among end points of the two reference contour line segments, and a gap exists between the two end points; in some embodiments, the distance between the two obstacle detouring points may be the minimum distance between the end points of the reference contour line segments where the two obstacle detouring points are respectively located, so as to judge the trafficability of the gap between the two obstacle detouring points.

As an embodiment, in the process that the robot walks along the obstacle detouring path, if the robot adjusts the walking direction of the robot to the direction from the current position to the predetermined target position and detects that the distance between the two obstacle detouring points is within the predetermined distance range, the robot determines that the robot starts to enter a narrow entrance, the narrow entrance is a gap formed between the obstacles where the two obstacle detouring points are respectively located, the gap is located in the passable area, the boundary line of the channel where the gap is located may be the contour line of the obstacle or the fitted contour line segment, and the distance between the boundary lines is within the predetermined distance range to improve the judgment accuracy. Preferably, the width of the notch is the distance between the grids where the two endpoints of the notch on the robot traveling plane are located, and the two endpoints of the notch are respectively located on the contour line of the corresponding obstacle or the fitted contour line segment. In addition, the robot configures the direction of the current position pointing to the set target position to be different from the reference obstacle detouring direction, optionally, the reference obstacle detouring direction corresponds to the arrow pointing from the last position of fig. 2 (which can be marked as the last obstacle detouring direction), and the direction of the current position pointing to the set target position corresponds to the arrow pointing from the next position of fig. 2 (which can be marked as the current obstacle detouring direction) and is all pointing to the passable area so that the robot avoids the surrounding obstacles; specifically, when the robot chooses to collide with an obstacle at one of the obstacle detouring points or walk to collide with the obstacle where one of the obstacle detouring points is located (not the obstacle detected in step S1), the robot adjusts its walking direction to the direction of the predetermined target position from the current position, and when the distance between the two obstacle detouring points is detected to be within the predetermined distance range, the gap formed between the obstacle currently collided with the robot and the obstacle detected in step S1 is regarded as a narrow road opening, and the robot is determined to start entering the narrow road opening, wherein the reference contour line segment is set to belong to the currently collided obstacle, the obstacle detouring path can belong to the historical obstacle detouring path, and the robot can determine the obstacle detouring direction of the robot according to the obstacle detouring path, the predetermined target position, the search start point in fig. 1 and 2, so as to choose to detouring the corresponding obstacle from the obstacle detouring point on the left side of the robot or the obstacle detouring point on the right side of the robot, and can twist left and right and touch the two obstacle detouring points successively. No matter how frequently the robot collides with the wound obstacle before entering the channel where the two obstacle-surrounding points are located, the walking direction of the robot is changed from the reference obstacle-surrounding direction to the direction from the current position to the preset target position (namely, the previous obstacle-surrounding direction is different from the current obstacle-surrounding direction), and when the distance between the two obstacle-surrounding points is in the preset distance range, the robot recognizes the channel where the two obstacle-surrounding points are located as a narrow channel, so that the robot can more smoothly navigate to the preset target position from the current position, the navigation path is shorter, and the navigation efficiency of the robot is improved.

In some embodiments, in the working area of the robot, there may be a working area framed by boundary lines, and a non-working area may be formed between the obstacles, or a non-working area may be formed between a corresponding boundary line of the two working areas, or a non-working area may be formed by a void channel between the two wall obstacles; the non-working area generally refers to a channel with a narrow width, namely a narrow channel for short, wherein an inlet of the narrow channel and an outlet of the narrow channel are both arranged as narrow channels, and two sides of the narrow channel are respectively provided with contour lines of two barriers; when the moving path is set, the sweeping robot is generally set as a point to be regarded as, and the set moving path may pass through a narrow channel in order to facilitate the sweeping robot to move to a target position; typically, a gap region between two or more obstacles (such as a channel between two walls) is displayed in the form of a free channel within the grid map; the free channel is used for communicating two different working areas and can be a passable road of the robot on the premise that the width is larger than the width of a body of the robot, then a narrow road opening and a narrow road allowing the robot to pass are identified by detecting the inlet of the channel, the outlet of the channel, the length of the channel and the width of the channel, and position information covered by the narrow road is determined, such as a grid area of a door opening below one wall body, a channel between two wall bodies and a channel with a gap to be identified formed by surrounding three wall bodies; the channel in the actual working scene of the robot has certain characteristics, including three-dimensional shape characteristics, dimension characteristics and the like.

As an embodiment, the obstacle detouring path in step S1 is a path planned by the robot using a heuristic search algorithm, and the predetermined target position is an end point of the obstacle detouring path, and the specific search start point may be a position point where the robot collides with the obstacle for the first time, or may be a position point selected near the obstacle, or may be the obstacle detouring point or a nearby area. Between the step S1 and the step S2, the method further includes: after the robot detects one obstacle (may collide with it), and before two obstacle detouring points are extracted in the search direction of the reference contour line segment of the other obstacle (the detected new obstacle may be relatively close in position to the previously detected obstacle, considered as being adjacent in position), the robot keeps walking along the obstacle detouring path, and sets the current extending direction of the obstacle detouring path as a history obstacle detouring direction, corresponding to the direction of the index number (3) pointed by the position where the index number (2) of fig. 2 is located, or the direction of the arrow from the last position of fig. 3.

Between the step S1 and the step S2, the method further includes: after the robot acquires the reference contour line segment in the process of walking along the obstacle detouring path, if the robot detects that the extending direction of the obstacle detouring path from the current position is not the direction from the current position to the predetermined target position target, namely, the direction of the arrow at the position where the index number (2) in fig. 2 is located is not directed to the predetermined target position target or the arrow from the last position in fig. 3 is not directed to the predetermined target position target, the robot stops to continue to walk along the obstacle detouring path, and adjusts the walking direction of the robot to the direction from the current position to the predetermined target position so that the robot starts to enter the channel where the two obstacle detouring points are located, specifically adjusting to: the direction of the arrow at the position of the index number (3) in fig. 3, or the direction of the arrow at the position of the index number (3) in fig. 2 (corresponding to the direction of the cur position in fig. 2 to the next position in fig. 2) so that the robot starts to enter the channel where the two obstacle detouring points are located, the robot continues to move straight toward the gap between the obstacles above and below in fig. 2 (such as to move along the direction of the index number (4) at the position of the index number (3) in fig. 2) or to move across the passable area at the line segment CD in fig. 3 to move along the direction of the next position at the position of the index number (3) in fig. 3 after recognizing that the behavior is entering the narrow road or recognizing the narrow road. Therefore, a narrower channel which is easy to misjudge and is formed between two obstacles is found out in the obstacle-detouring walking process of the robot, the direction is adjusted to enter the channel, the robot does not continue to detour, the whole navigation path of the robot is shortened, and the navigation to the set target position is quickened.

As an example, as can be seen in conjunction with fig. 2 and 3, the present example marks the obstacle above fig. 2 and 3 as a second obstacle, and the obstacle below fig. 2 as a first obstacle; in fig. 3, the reference contour line segment AB of the second obstacle has a first end point a and a second end point B, which are located on the left and right sides of the lower left corner point of the second obstacle (black filled rectangle shown in fig. 3). In this embodiment, when the robot walks along a path pre-planned by the heuristic search algorithm, the robot may collide with the first obstacle when using a ranging sensor (such as a laser sensor) at a position start to detect the first obstacle, and simultaneously, the robot plans an obstacle detouring path leading to a predetermined target position, specifically, uses the collision position as a search starting point and the predetermined target position as a search end point, and uses heuristic search algorithm rules to mark an obstacle detouring path; then the robot walks along the obstacle detouring path to walk along the contour line of the first obstacle until the robot walks to the notch formed between the first obstacle and the second obstacle, corresponding to fig. 2, the robot walks to the position of the index number (2) along the arrow direction of the position of the index number (1) from the start position, and walks to the position of the index number (3) along the arrow direction of the position of the index number (2), the robot walks to the notch formed between the first obstacle and the second obstacle, corresponding to the cur position of fig. 3, and at the moment, the robot can recognize the channel where the two obstacle detouring points are located as a narrow channel when the robot detects that the distance between the two obstacle detouring points (which can correspond to the point C and the point D of fig. 3) is in the preset distance range. Meanwhile, the direction formed by the robot in the walking process is marked as the history obstacle detouring direction, and the direction corresponding to the arrow point of the position where the index number (2) is located; as shown in fig. 3, the robot acquires a reference contour line segment AB of the second obstacle, marks the direction of the first end point a pointing to the second end point B as a direction of pointing to a set target position from the current position (the direction of an arrow corresponding to the position of the index number (3)) so as to pass through the channel where the two obstacle detouring points are located, marks the direction of the second end point B pointing to the first end point a as an extending direction of the history obstacle detouring direction, and can detouring the contour above the second obstacle in the process of walking along the extending direction of the history obstacle detouring direction, thereby detouring to the upper side of the obstacle and navigating to the set target position target, and realizing that the robot turns to walk along the contour line of the second obstacle after detouring the first obstacle, so as to keep executing the obstacle detouring operation; corresponding to the previous embodiment, the robot configures an extension direction of the historical obstacle detouring direction as the reference obstacle detouring direction to represent a general trend of the upper left contour line of the first obstacle and the second obstacle.

It should be added that, for the aforementioned narrow road, narrow road or gap, in the area where the robot is located, there is a gap between two parallel walls and a door opening penetrating through two adjacent room areas on the horizontal ground, the entrance of the door opening or the entrance of the gap can be set as an opening formed between obstacles, which are openings formed between the outlines of at least two obstacles, and these obstacles can exist in isolation from each other; when the robot scans the surrounding environment by using the laser sensor, both end points and the width of the opening are scanned out by the laser sensor of the robot and converted to point cloud coordinates under a corresponding coordinate system, so that both end points of the opening are scanned into corresponding point clouds and converted to corresponding grids of the grid map. In some embodiments, when the robot senses the surrounding environment by collision, each time the collision sensor of the robot contacts two end points of the opening, the contour points of the obstacle or the reference contour line segments on which the collision sensor of the robot collides are marked at the corresponding grids of the grid map; in some implementations, an evaluation is also introduced to indicate the degree of trafficability of the robot at the gap formed between obstacles or at its corresponding grid (the number of free grids between two end points or its duty ratio in the range of the opening width), which may indicate the probability of traffic at the corresponding grid area, typically the reliability that the robot is given when scanning out the gap using the ranging sensor is higher than the reliability that the robot is given when collision detects the gap, because the positioning accuracy of the ranging sensor is higher than the positioning accuracy that would result from physical contact of the robot.

Preferably, the extending direction of the passable area in the channel where the two obstacle detouring points are located is parallel to the reference contour line segment (corresponding to the line segment AB in fig. 3) of the second obstacle, the reference contour line segment of the second obstacle is perpendicular to the searching direction (corresponding to the direction of the point D pointing to the point C in fig. 3), the passable area in the channel where the two obstacle detouring points are located is communicated with the predetermined target position, one obstacle detouring point is located on the reference contour line segment of the second obstacle, for example, the point D and the point C are marked as the obstacle detouring point of the second obstacle above and the obstacle detouring point of the first obstacle below, respectively, the point D is located on the reference contour line segment AB, the area where the line segment CD is located is the passable area, and the line segment CD can be marked as the narrow crossing.

In the foregoing embodiment, the robot collects point cloud data by the ranging sensor, where the point cloud data is configured to reflect the position information of the obstacle detected by the ranging sensor, and is a set of a series of discrete points, and may carry environmental noise (feedback of the influence of the environmental light interference or the obstacle indicating material); the robot then fits the acquired point cloud data into the contour line of the obstacle to represent the local contour of the detected obstacle or the obstacle envelope; the fitting processing tool is sequentially subjected to sorting, grouping (to distinguish different types of obstacles), screening, piecewise interpolation fitting, and fitting curve coordinate points on each group of connecting lines to obtain an obstacle envelope line or an obstacle contour line, specifically a fitting line segment, a fitting curve and a combination thereof, wherein the reference contour line segment belongs to the contour line obtained by fitting processing. Preferably, in order to identify a narrow lane with a passable meaning, when the robot marks the extending direction of a passable area inside a lane where two obstacle detouring points are located as a preset lane direction, the two obstacle detouring points are respectively located on a contour line of a first obstacle in the preset lane direction and a contour line of a second obstacle in the preset lane direction, or the two obstacle detouring points are respectively located on the first obstacle and the second obstacle, wherein a distance between the two obstacle detouring points is set to be the preset distance range so as to be convenient for identifying that the robot enters the narrow lane, or a width of the lane where the two obstacle detouring points are located is located to be the preset distance range so as to be convenient for determining that the robot can pass through the narrow lane; the preset channel direction is the direction in which the index number (3) in fig. 3 points to the next position, or the direction in which the index number (3) in fig. 2 points to the index number (4), so that the entrance of a narrower small channel and the way of entering a narrow channel are formed in the obstacle-surrounding walking process of the robot.

As an example, as can be seen from fig. 2 and 3, when the robot detects the first obstacle, the position along the obstacle detouring path and the predetermined target position are located on both sides of the first obstacle, for example, the start position and the target position in fig. 2 are located on both sides of the first obstacle; then the robot marks the side where it is located along the obstacle detouring path as a first side, corresponds to the left side of the first obstacle of fig. 2 and 3, and marks the side where the given target position target is located as a second side, corresponds to the right side of the first obstacle of fig. 2 and 3, and configures the first end point a of the reference contour line segment of the second obstacle to be located at the first side of the second obstacle or the first side of the first obstacle, and configures the second end point B of the reference contour line segment of the first obstacle to be located at the second side of the second obstacle or the second side of the first obstacle; the first barrier and the second barrier are respectively surrounded by the inner side and the outer side, the inner side is the right side, the outer side is the left side, the extending direction of the narrow channel is the right side, and the inlet of the narrow channel is the left side. And then the robot marks the profile line trend formed by connecting the profile line segment on the first side of the first obstacle and the profile line segment on the first side of the second obstacle as the outer profile line trend formed by the obstacles where the two obstacle detouring points are respectively located, and the outer profile line trend corresponds to the reference obstacle detouring direction disclosed in the previous embodiment, so as to represent the extending directions of the profile lines which are planned under the condition of heuristic search algorithm and are close to the upper left side of the first obstacle and the upper left side of the second obstacle.

Preferably, the robot has collided with the first obstacle when the robot detects the first obstacle using the laser sensor; when the robot walks to a gap formed between the first obstacle and the second obstacle, the robot collides with the second obstacle; and the robot sets the preset distance range to be larger than or equal to the width of the robot body, so that when the distance between the two obstacle detouring points is larger than or equal to the width of the robot body, the gap formed between the obstacles where the two obstacle detouring points are respectively positioned is determined to be a narrow opening for allowing the robot to enter. The upper limit value of the preset distance range is the sum of the body width of the robot and the preset redundancy. The width of the narrow channel is only slightly larger than the width of the robot body, and a preset distance range is set for judgment in order to reduce error interference of the calculated contour lines in the process of identifying the narrow channel; specifically, the minimum value (lower limit value) of the preset distance range is larger than the body width of the robot, and the maximum value (upper limit value) of the preset distance range is only slightly larger than the body width of the robot and does not exceed twice the body width. The preset distance range can be determined according to the width of the robot body and a preset fitting error, and the setting of the preset redundancy amount can refer to a channel which is relatively close to the width of the robot body as the narrow channel. For example, the width of the robot for sweeping floor is 30 cm, the preset distance may be 32 cm to 35 cm, that is, when the center of the robot is located at the center position of the entrance (or the notch, such as the midpoint of the line segment CD in fig. 3) of the channel where the two obstacle detouring points are located in the width direction, the preset redundancy is set as the gap between the left and right sides of the robot and the channel is between 1 cm and 2 cm.

In summary, the foregoing embodiment uses the change of the obstacle winding direction before entering the gap formed between two wound obstacles and the distance between the obstacle winding points corresponding to the two wound obstacles to identify whether the robot starts to enter the narrow lane, so as to overcome the error brought to the narrow lane identification by the contour line fitting calculation of the obstacles, and improve the accuracy of narrow lane identification and the success rate of the robot entering the narrow lane.

Preferably, the narrow channel is a channel composed of boundary lines of contour lines (fitting results) of the two obstacles; the minimum distance between the obstacle detouring points extracted from the contour lines of the two obstacles is larger than or equal to the minimum value of the preset distance range; the minimum value of the preset distance range is larger than the width of the robot body; the width of the narrow channel is in a preset distance range, and the entrance of the narrow channel is also in the preset distance range. The narrow road junction can be a small door opening of a room, the barriers at two sides of the door opening are four walls in the same room, the four walls are continuous and integrated, and in addition, the walls at the left side and the right side of a channel where the barrier-surrounding points are located can be approximately parallel.

Preferably, when the robot walks inside the narrow channel, the shortest distance between the boundary lines at two sides of the entrance of the narrow channel and the corresponding side of the robot is equal to half of the preset redundancy amount; for example, when the robot enters the narrow road junction along the center line of the narrow road, the vertical distance between the left side of the robot and the obstacle-surrounding point on the left side of the robot is equal to half of the preset redundancy, and the vertical distance between the right side of the robot and the obstacle-surrounding point on the right side of the robot is also equal to half of the preset redundancy. In this embodiment, half of the preset redundancy is preferably 1 cm to 2 cm, so as to avoid the fitting calculation error of the contour line.

Based on the foregoing embodiment, the present invention also discloses a chip storing program codes which when executed implement the steps of the method for identifying the robot entering the narrow channel as described. When the program code corresponding to the steps of the method for identifying the robot entering the narrow channel is stored in a chip, the program code is used as a computer program product, and the computer program is operable to cause a computer to execute part or all of the steps of any one of the methods described in the embodiment of the method for identifying the robot entering the narrow channel. The robot with the chip inside obtains the outline line segment by utilizing the laser data and extracts relevant obstacle detouring points in the obstacle detouring walking process, identifies the narrow channel with the width slightly larger than the width of the robot body and determines that the robot can enter the narrow channel, and is also convenient for screening a more smooth navigation path.

The invention also discloses a robot, which is provided with the ranging sensor, and the robot is provided with the chip disclosed in the embodiment, so that the robot is controlled to detect the obstacle by using the ranging sensor and obtain the corresponding contour line and the obstacle detouring point, and the robot can be conveniently identified to start entering the narrow road. When the robot walks along the obstacle detouring path planned in advance, the robot aims at the problem that a narrow road formed between two obstacles and just larger than the width of a machine body by a small clearance distance (1 cm) is easy to misjudge as non-passing, the contour line segment related to the passing factor of a gap formed between the two obstacles is obtained and used for guiding the robot to adjust the obstacle detouring direction before and after entering the gap, and when the obstacle detouring direction is changed, the channel where the two obstacle detouring points are located is identified as the narrow road by combining the distance between the two obstacle detouring points extracted in the searching direction of the contour line segment, and the robot is determined to start entering a narrow road port but not enter the obstacle detouring path planned in advance. Therefore, the narrow channel for the robot to pass is accurately distinguished in the environment where the sweeping robot, the mower or the mobile toy is located, the problem that the width of the narrow channel is small enough to easily misjudge the grid area inside the narrow channel or the opening of the narrow channel as an obstacle blocking the robot to pass is solved, smooth switching from the action of bypassing the obstacle by the robot to the action of entering the narrow channel formed between two obstacles is also realized, and the robot can accurately pass through the narrow channel or the narrow channel according to laser data, so that the smoothness and success rate of passing through the narrow area under the laser navigation condition are improved.

It should be noted that, for simplicity of description, the foregoing method embodiments are all expressed as a series of action combinations, but it should be understood by those skilled in the art that the present application is not limited by the order of actions described, as some steps may be performed in other order or simultaneously in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required in the present application.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, such as the division of the units, merely a logical function division, and there may be additional manners of dividing the actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, or may be in electrical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a memory, including several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application. And the aforementioned memory includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Those of ordinary skill in the art will appreciate that all or a portion of the steps in the various methods of the above embodiments may be implemented by a program that instructs associated hardware, and the program may be stored in a computer readable memory, which may include: flash disk, read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), magnetic or optical disk, and the like.

The foregoing has outlined some of the embodiments of the present application in detail, and the detailed description of the principles and embodiments of the present application has been provided herein by way of example only to facilitate the understanding of the method of the present application and the core concepts thereof; meanwhile, as those skilled in the art will have modifications in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. An identification method for a robot entering a narrow channel, wherein an execution main body of the identification method is a robot provided with a ranging sensor; the identification method is characterized by comprising the following steps:

step S1, when a robot detects an obstacle by using a ranging sensor, the robot plans an obstacle detouring path leading to a set target position, walks along the obstacle detouring path, and acquires a reference contour line segment through point cloud data acquired by the ranging sensor;

And S2, under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the robot detects that the walking direction of the robot changes from the reference obstacle detouring direction to the direction pointing to the preset target position from the current position and detects that the distance between the two obstacle detouring points is in the preset distance range, the robot recognizes the behavior as entering a narrow road.

2. The identification method according to claim 1, wherein the step S2 further comprises:

under the condition that the robot extracts two obstacle detouring points in the searching direction of the reference contour line segment, if the distance between the two obstacle detouring points is detected to be in a preset distance range, identifying a channel in which the two obstacle detouring points are positioned as a narrow channel;

the direction of the current position pointing to the set target position in step S2 is pointing to the passable area inside the channel where the two obstacle detouring points are located.

3. The identification method according to claim 2, wherein the robot configures the reference obstacle detouring direction as an outer profile trend of an obstacle where each of the two obstacle detouring points is located, so as to form an extending direction of the obstacle detouring path before the robot enters the narrow road;

the robot sets the searching direction of the reference contour line segment as the vertical direction of the reference contour line segment, and forms the width direction of the channel where the two obstacle detouring points are located; the reference contour line segment corresponds to an obstacle where one obstacle detouring point is located, and forms one boundary line of the channel where the two obstacle detouring points are located.

4. A method of identifying as claimed in claim 3, wherein during the course of the robot's travel along the obstacle detouring path, if the robot adjusts its travel direction to a direction from the current position to the intended target position and detects that the distance between the two obstacle detouring points is within the predetermined distance range, the robot determines that it is starting to enter a narrow gap formed between the obstacles in which the two obstacle detouring points are each located, the gap being located in the passable area, the robot further configuring the direction from the current position to the intended target position to be different from the reference obstacle detouring direction.

5. The identification method according to claim 2, characterized in that between said step S1 and said step S2, further comprising:

after the robot detects one obstacle and before two obstacle detouring points are extracted in the searching direction of the reference contour line segment of the other obstacle, the robot keeps walking along the obstacle detouring path, and the current extending direction of the obstacle detouring path is set as a historical obstacle detouring direction;

and after the robot acquires the reference contour line segment in the process of walking along the obstacle detouring path, if the robot detects that the extending direction of the obstacle detouring path from the current position is not the direction from the current position to the set target position, stopping to walk along the obstacle detouring path, and adjusting the walking direction to the direction from the current position to the set target position so as to enable the robot to start entering the channel where the two obstacle detouring points are located.

6. The method according to claim 5, wherein when the robot detects the first obstacle using the ranging sensor, the robot plans an obstacle detouring path leading to a predetermined target position, and then walks along the obstacle detouring path to achieve walking along the contour line of the first obstacle until the robot marks the direction of walking to the gap formed between the first obstacle and the second obstacle and/or collides with the second obstacle as the historical obstacle detouring direction, and a reference contour line of the second obstacle has a first end point and a second end point; then the robot acquires a reference contour line segment of the second obstacle, marks the direction of the first end point pointing to the second end point as a direction pointing to a set target position from the current position, and marks the direction of the second end point pointing to the first end point as an extending direction of the history obstacle detouring direction; the robot configures an extension direction of the history obstacle detouring direction as the reference obstacle detouring direction.

7. The method according to claim 6, wherein the extending direction of the passable area inside the passage where the two obstacle detouring points are located is parallel to the reference contour line of the second obstacle, the reference contour line of the second obstacle is perpendicular to the searching direction, the passable area inside the passage where the two obstacle detouring points are located is communicated with the predetermined target position, and one obstacle detouring point is located on the reference contour line of the second obstacle.

8. The recognition method according to claim 7, wherein the robot collects point cloud data through a ranging sensor, the point cloud data being position information configured to reflect an obstacle detected from the ranging sensor; the robot then fits the acquired point cloud data into the contour line of the obstacle to represent the local contour of the detected obstacle or the obstacle envelope;

the reference contour line segment belongs to a contour line which is subjected to fitting processing;

when the robot marks the extending direction of the passable area in the channel where the two obstacle detouring points are located as the preset channel direction, the two obstacle detouring points are respectively positioned on the contour line of the first obstacle in the preset channel direction and the contour line of the second obstacle in the preset channel direction, or the two obstacle detouring points are respectively positioned on the first obstacle and the second obstacle.

9. The recognition method according to claim 8, wherein when the robot detects the first obstacle, the position along the obstacle detouring path and the predetermined target position are separated on both sides of the first obstacle, then the robot marks a side where the position along the obstacle detouring path is located as a first side, marks a side where the predetermined target position is located as a second side, and the robot configures a first end point of a reference contour line segment of the second obstacle to be located on the first side of the second obstacle or the first side of the first obstacle, and the robot configures a second end point of the reference contour line segment of the first obstacle to be located on the second side of the second obstacle or the second side of the first obstacle;

And then the robot marks the profile line trend formed by connecting the profile line of the first side of the first obstacle and the profile line of the first side of the second obstacle as the profile line trend of the outer profile line formed by the obstacles where the two obstacle detouring points are respectively positioned.

10. The method of claim 6, wherein the robot has collided with the first obstacle when the robot first detected the first obstacle using the ranging sensor; when the robot walks to a gap formed between the first obstacle and the second obstacle, the robot collides with the second obstacle;

the robot sets the preset distance range to be greater than or equal to the body width of the robot, and the upper limit value of the preset distance range is the sum of the body width of the robot and the preset redundancy amount.

11. A chip storing program code, characterized in that the program code when executed realizes the steps of the method for identifying entry of a robot into a narrow lane according to any one of claims 1 to 10.

12. A robot provided with a ranging sensor, characterized in that the robot is provided with the chip of claim 11, and the chip is used for controlling the robot to detect an obstacle by using the ranging sensor and obtain a corresponding contour line and an obstacle detouring point, so that the robot can be conveniently identified to start entering a narrow channel.